1. Field of the Invention
The present invention relates to midecamycin biosynthesis genes which are involved in the production of midecamycins, and more specifically to genes encoding functional modules of polyketide synthases.
2. Background Technology
Since macrolide antibiotics which are effective to gram-positive bacteria, mycoplasms, chlamydias and the like can be orally administered and have low toxicity, they are classified as clinically important antibiotics. In particular, commercially-available 16-membered ring macrolide antibiotics are widely used in the world, mainly in Asian countries, because of their advantages, for example, that they are less likely to induce resistant strains and less interactive with other drugs than 14-membered ring macrolides, and have little effect on the intestinal tract.
Midecamycins (FIG. 1) belong to 16-membered ring macrolide antibiotics and several analogues have been reported. They are clinically used extensively along with miokamycin, an acylated derivative of a midecamycin (Omoto, S. et al., J. Antibiot., 29, 536 (1976); Yoshida, T. et al., Jpn. J. Antibiot., 35, 1462 (1982)).
Midecamycins are produced by a species of actinomycetes, Streptomyces mycarofaciens (ATCC 21454), and industrial scale production by fermentation using this strain has been established. Conventionally, actinomycetes have an important role in the field of fermentation industry as microorganisms for the production of secondary metabolic products, such as antibiotics and physiologically active substances, and their productivity has been improved by various microbial breeding techniques. The microbial breeding has also been carried out for midecamycin production by Streptomyces mycarofaciens by inducing mutation with various mutagens.
Recently, recombinant DNA technology has been introduced to improve productivity of secondary metabolites and to create novel active substances and a number of genes in secondary metabolic systems have already been isolated. Examples of isolated genes involved in the production of macrolide antibiotics include tylosin biosynthesis genes (Merson-Davies, L. A. and Cundliffe, E., Mol. Microbiol., 13, 349 (1994); Gandecha, A. R. et al., Gene, 184, 197 (1997); Wilson, V. T. and Cundliffe, E., Gene, 214, 95 (1998); Fouces, R. et al., Microbiology, 145, 855 (1999); Bate, N. et al., Microbiology, 146, 139 (2000); Review: Cundliffe, E. et al., Antonie Van Leeuwenhoek, 79, 229 (2001); U.S. Pat. No. 5,876,991, U.S. Pat. No. 5,672,497, U.S. Pat. No. 5,149,638, European Patent No. 791655, European Patent No. 238323), nidamycin biosynthesis genes (Kakavas, S. J. et al., J. Bacteriol., 179, 7515 (1997); WO98/51695), and erythromycin biosynthesis genes (Dhillon, N. et al., Mol. Microbiol., 3, 1405 (1989); Cortes, J. et al., Nature, 348, 176 (1990); Donadio, S. et al., Science, 252, 675 (1991); Haydock, S. F. et al., Mol. Gen. Genet., 230, 120 (1991); Stassi, D. et al., J. Bacteriol., 175, 182 (1993); Linton, K. J. et al., Gene, 153, 33 (1995); Gaisser, S. et al., Mol. Gen. Genet., 256, 239 (1997); Summers, R. G. et al., Microbiology, 143, 3251 (1997); Gaisser, S. et al., Mol. Gen. Genet., 258, 78 (1998); Salah-Bey, K. et al., Mol. Gen. Genet., 257, 542 (1998); WO93/13663, U.S. Pat. No. 6,004,787, U.S. Pat. No. 5,824,513, WO97/23630, U.S. Pat. No. 5,998,194).
In microorganisms which produce macrolide antibiotics, most of the macrolide biosynthesis genes are often clustered together in a region of 70 to 80 kb in the genome (Donadio, S. et al., Science, 252, 675 (1991); MacNeil, D. J. et al., Gene, 115, 119 (1992); Schwecke, T. et al., Proc. Natl. Acad. Sci., 92, 7839 (1995)). In the center of such clusters, there exists a highly homologous gene called Type I polyketide synthase (PKS) which encodes a huge multi-functional protein.
The PKS is generally composed of 3 to 5 genes and its protein forms a complex comprising an initiator module and several extender modules. Each of these components adds a specific acyl-CoA precursor to a polyketide chain in the process of synthesis to specifically modify β-keto groups. Accordingly, the structure of polyketide is determined by the composition and the order of these modules in the PKS. The modules contain several domains and each of them has its specific function.
The initiator module is composed of an acyl-carrier protein (ACP) domain to which an acyl group of precursor binds and an acyltransferase (AT) domain which catalyzes addition of the acyl group to the ACP domain. Difference in specificity of this AT domain determines the kind of acyl-CoA to be added thereto. All of the extender modules contain a β-ketosynthase (KS) domain, which adds a previously existing polyketide chain to a new acyl-ACP by decarboxylation condensation, the AT domain and the ACP domain.
Further, in addition to these domains, the extender modules contain several domains which modify specific β-keto groups and the composition of the domains contained determines the modification of β-keto groups. Such domains include a β-ketoreductase (KR) domain which reduces a β-keto group to a hydroxyl group, a dehydratase (DH) domain which removes a dehydroxyl group and generates a double bond, and an enoylreductase (ER) domain which reduces a double bond and generates a saturated carbon bond.
The last extender module ends with a thioesterase (TE) domain which catalyzes the cyclization and release of polyketide from the PKS.
A polyketide skeleton produced by PKS undergoes further modifications, such as methylation, acylation, oxidation, reduction, and addition of specific sugars, to ultimately synthesize macrolide antibiotics. Most of the genes necessary for these modifications exist in the vicinity of the PKS gene.
As for genes involved in midecamycin biosynthesis, a midecamycin self-resistance gene (mdmA; Hara, O. and Hutchinson, C. R., J. Antibiot., 43, 977 (1990)), a 3-O-acyltransferase gene (mdmB), an O-methyltransferase gene (mdmc; Hara, O. and Hutchinson, C. R., J. Bacteriol., 174, 5141 (1992)), and a 4″-O-propionyltransferase gene (mpt; Xulun, Z. and Yiguang, W., ActaMicrobiol. Sci., 36, 417 (1996)) have been reported. However, no other gene involved in midecamycin biosynthesis has been reported.